Semiconductor industry has been developed markedly by miniaturizing devices with an improvement in recent semiconductor microprocessing technology. The microprocessing technology typified by lithography has recently achieved an accuracy of finishing of several hundred nanometers. Materials and devices applied with the above technology can participate actively in many scenes and have been expected to be applied in a wide variety of fields of optical communication, electric communication, and so on as well as of biotechnology and energy. However, when consideration is given to processing on the scale of 100 nanometers or less as an extension of current microprocessing technology, there remain many challenges for industrial use including the time and expense involved in processing in addition to technical challenges. With a growing number of applicable fields, novel technology of manufacturing refined structures as an alternative to the above has been strongly demanded.
Under the circumstances, research and development have been actively conducted on novel materials by means of a bottom-up approach that produces desired structures and properties controlled at the atomic or molecular level, instead of conventional top-down processing technology. An example of bottom-up technology of producing refined structures includes the development of a molecular device in which a molecular arrangement is controlled on a metal surface, and the use of methods of utilizing the self-assembly of substances has been studied as one of such technologies of molecular arrangement control. As a recent example of the study, Lindsay et al., (Science, 300: p. 1413, 2003) has examined a molecular switching function by the use of a self-assembled monolayer where alkanethiol and alkanedithiol having thiol groups at their respective molecular ends are oriented on a gold substrate.
It has been known that biomolecules typified by nucleic acids and proteins are constructed as precise structures under control at the atomic level for exerting their functions. The uses of the properties of such biomolecules have been also under investigation to apply the biomolecules to a variety of devices in which the biomolecules are arranged on a substrate of metal, metal oxide, or semiconductor. Technology of producing a fine structure where the biomolecule is arranged on a substrate becomes important for a first step of developing such devices. For example, when consideration is given to the immobilization of a deoxyribonucleic acid (DNA) on a gold substrate or the immobilization of a peptide having an amino acid sequence with various functions thereon, it has been widely known that the DNA or the peptide can be chemically synthesized and the ends of these substances are chemically modified with thiol groups (—SH) to thereby coordinate these substances on a gold substrate with the use of S—Au surface absorption. Using this fact, study has been conducted on experimental systems and conversion into devices in which DNA or peptide is immobilized on gold.
On the other hand, proteins that function as enzymes, antibodies, or the like have high molecular weights. Thus, it is very difficult to chemically synthesize such compounds having high molecular weights to form higher order structures while keeping abilities to exert their functions. Although a variety of studies have been currently conducted, the situation is that functional proteins that combine their higher order structures with desired functions have not been synthesized yet in most cases (Science, 302: p. 1364, 2003). When the functional protein is immobilized on a substrate material, binding with the substrate is generally performed by treating the substrate with a variety of surface treatment agents typified by a silane coupling agent, and introducing into the protein a functional group capable of binding to the surface of the substrate that was subjected to surface treatment (Proteomics, 3: p. 254, 2003). However, it has been pointed out that such introduction of a reactive functional group into the functional protein is generally performed by chemical modification and an introduction site is nonselectively determined. Therefore, the functional protein is immobilized in a less functionable shape on the substrate and may be reduced in the resulting activity due to the modification of the functional expression site of the functional protein.
It is also possible to introduce a binding site into a protein by a genetic engineering procedure to produce a fusion protein. As an example, there is known a method in which all sites or the biotin-binding site of (strept) avidin which is known to bind to a low molecular compound, biotine, is introduced into the N- or C-terminal of a desired protein by genetic engineering and the desired protein is then expressed as a fusion protein that is in turn immobilized via biotin arranged on the surface of a desired substrate.
Furthermore, another technique has been recently disclosed. In this case, a peptide composed of five or more amino acids capable of binding to a substrate material for immobilization. The peptide may be fused to a desired protein to produce a fusion protein, thereby binding and immobilizing the desired protein on the substrate. Belcher et al., has disclosed a peptide composed of 12 amino acids capable of specifically recognizing a certain crystal surface containing GaAs by the phage display method (Nature, 405: p. 665, 2000), creating the new possibilities for studying devices with the use of the ability of biomaterials to self-assemble. Moreover, Belcher et al., discloses the amino acid sequence of a peptide composed of 7 to 14 amino acids for other semiconductors (PbS, CdS) (WO 03/029431). Brown et al., discloses some examples of a peptide with a repetitive structure having the unit of an amino acid sequence composed of 14 residues, showing affinity for gold (Nature Biotechnology, 15: p. 269, 1997). Heretofore, a number of other peptides having affinity for metals (Au, Pt, Pd, Ag), metal oxides (SiO2, ZnO, Cr2O3, Fe2O3), or semiconductors have been obtained by the above-described phage display method.
By immobilizing a desired biomolecule on a substrate via such a peptide having affinity for the substrate material, a structure cyclically having a desired function and shape can be also constructed in a self-assembly manner due to the interaction between biomolecules or with a different substance. For example, Belcher et al., discloses technique in which M13 phage displaying ZnS-affinity peptide bound with ZnS particles is coordinated by self-assembly to thereby produce a liquid crystal-like film (Science, 296: 892, 2002).
In addition, some examples show a desired functional protein to which a substrate-binding site is fused is produced by genetic engineering as described above to thereby allow the functional protein to be immobilized at a desired site other than the functional site on the substrate. However, when peptide having affinity for a substrate material is lined to the end of a functional protein directly or via a linker composed of a few amino acids and then bound to a solid phase (e.g., substrate) to achieve the immobilization of the protein, the active site of the functional protein (e.g., several amino acid residues constituting the antigen-binding site of an antibody) is close to the substrate and undergoes some interaction (e.g., electrostatic effect) from the surface of the substrate to cause structural changes and the like, resulting in a fear of incapable of sufficiently exerting a desired function. Even when a linker is designed to be still longer, peptide as the linker, which has a high degree of freedom in the motion of the molecules, may be arranged close to a functional site and its function may be inhibited.
From the above problems, the inventors of the present invention have arrived at the idea that a substrate-binding site on which functional polymeric materials including a functional protein linker are immobilized needs to have a structural portion (scaffold) given as a spacer that keeps a certain distance from a substrate without inhibiting desired activity of an immobilized protein, and a site that binds to the substrate.
As a molecule-recognizing molecule having such a scaffold, the most well known is antibody. The antibody is one of the proteins functioning in a self defense mechanism to specifically recognize and bind to a variety of structures on the surface of a foreign substance invading animals body fluid that is subsequently detoxicated by the immune system. The diversity of antibodies (the number of antibodies having different amino acid sequences for binding to a variety of foreign substances) is estimated to be 107 to 108 different varieties per animal. The structure is formed of polypeptide chains having two longer chains and two shorter chains and the longer polypeptide chain (heavy chain) and the shorter polypeptide chain (light chain) are referred to as a heavy chain and a light chain, respectively.
Those heavy and light chains each have a variable region and a constant region. The light chain is a polypeptide chain composed of two domains of one variable region (VL) and one constant region (CL), while the heavy chain is a polypeptide chain composed of four domains of one variable region (VH) and three constant regions (CH1-CH3). Each of the above-described domains assumes a cylindrical structure composed of approximately 110 amino acids, and forms the layer structure of β sheets antiparallely arranged, and this layer structure is bound with one SS bond to form a highly stable structure. Moreover, it is known that the binding of the antibody to a variety of antigen species results from the diversity of amino acid sequences in three complementarity determining regions (CDF) of each of the above-described variable regions (VH or VL). The CDRs, three for VH and three for VL, are separately arranged by framework regions and recognizes the spatial configuration of functional groups in a recognition site of interest to allow highly specific molecular recognition.
The diversity of the above-described CDR is attributed to DNA rearrangement that takes place in the antibody locus when a myeloid stem cell is differentiated into a B lymphocyte as an antibody-producing cell. It is known that a portion composed of VH, D, and JH gene fragments in the heavy chain, and a portion composed of Vλ or Vκ and Jλ or Jκ gene fragments in the light chain undergo DNA rearrangement to thereby produce antibodies. Such DNA rearrangement independently takes place in each B cell, one B cell produces only one kind of antibody. However, the whole B cells in an individual can produce diverse antibodies.
Antibody capable of binding to a specific substance as described above has been artificially produced heretofore using an antibody-forming mechanism in the immune system of an animal, and utilized in a variety of industrial fields. An example of production methods is a method in which an animal to be immunized (e.g., rabbit, goat, or mouse) is immunized at constant intervals with an antigenic substance of interest along with an adjuvant, and antibodies present in its serum are collected. The antibodies obtained as above are a mixture of several antibodies recognizing a variety of structures in the surface of the antigenic substance used in the immunization. Serum containing several antibodies binding to one antigen is referred to as a polyclonal antibody.
On the other hand, a variety of B lymphocytes that produce antibodies binding to an antigen of interest are present in the spleen of an animal to be immunized. Such an antibody-forming B lymphocyte is fused to an established tumor cell to thereby allow the production of a hybridoma cell. As described above, one B lymphocyte produces one kind of antibody, and a system that can subculture the B lymphocyte producing one kind of antibody as a hybridoma has been established. The antibody produced as above is referred to as a monoclonal antibody.
As a molecular recognition structure using the above-described structure of the antibody as an anchorage, JP 05-055534 A discloses a multibinding antibody recognizing two different antigens and a method of forming a multilayer using the same. According to this technique, a fusion antibody recognizing a first antigen and a second antigen can be obtained. In addition, the first antigen, the fusion antigen, and the second antigen are successively arranged on the surface of a substrate to thereby allow the formation of a multilayer without chemical modification. However, for obtaining the disclosed fusion protein, it is necessary to use an animal cell, turning into a problem in terms of cost efficiency and operational complication. Besides, when the multilayer is formed, a step of arranging an antibody recognizable by the fusion protein on the surface of the substrate must be provided.
An antibody fragment, Fab, Fab′, or F(ab′)2, obtained by treating the above-described antibody with a certain proteolytic enzyme, is known to have the ability to bind to an antigen similar to that of the parent antibody.
Likewise, for the above-described VH, VL, or Fv as a complex thereof, or even for a complex composed of the VH or the VL, single chain Fv (scFv) having the carboxy terminal of one region and the amino terminal of another region linked via peptide composed of several amino acids, or the like, is known to have the ability to bind to an antigen similar to that of the parent antibody.
Skerra and Better disclose a method in which Fab-type and Fv-type antibody fragments, the N terminals of which are added with a secretion signal sequence by a genetic engineering method, are produced as the expression of antibody genes by E. coli (Science, 240: P. 1038, 1988; Science, 240: p. 1041, 1988). Alternatively, JP 07-501451 A discloses a multivalent antigen-binding antibody and a method of producing the same, and JP 08-504100 A discloses a multivalent and multibinding antibody and a method of producing the same. Those two techniques disclose that a binding protein is a protein containing an antibody variable region portion (VH and/or VL) binding to one or more antigens. JP 07-501451 discloses the structures and amino acid sequences of binding proteins each recognizing pancarcinoma antigen TAG-72 and fluorescein and a bispecific antibody recognizing both, as well as base sequences encoding them. JP 08-504100 A discloses, in Examples, a divalent and bispecific binding protein composed of an antibody fragment complex for known proteins (cell membrane protein, cancer antigen CEA, FcRγ1, etc.,) or low molecular compounds.
However, in the above-described disclosures, no technique concerning a binding protein that directly recognizes and binds to the surface of a substrate material typified by an inorganic substance is disclosed. Therefore, when a structure in which the binding protein is arranged on the substrate is produced, the arrangement must be performed by a conventionally known method, for example, a method involving chemically modifying the substrate or the above-described binding protein to arrange the binding protein on the substrate through a covalent bond. Similarly, when the binding protein is bound to a fine particle of a metal or a semiconductor substance, it is also necessary to chemically modify the particle to be bound or binding protein. Such chemical modification often targets amino residues or carboxyl groups contained in large numbers in the molecule of a protein, and a site involved in the reaction is nonselective. Therefore, a site exerting desired activity may be a substrate-binding or labeling site and, as a result, the desired activity of the protein may be reduced. Moreover, such a problem is likely to arise in not only the technique of conversion into microdevices but the production of sensing elements such as biosensors. Thus, it is important to immobilize a molecule capturing a target substance such as an antibody on a substrate while molecular orientation that sufficiently exerts its capturing function is maintained.
Studies are being conducted, in which a protein having a scaffold similar to that of the above-described antibody is added with the ability to recognize diverse molecules. Examples thereof include anticolin (Review in Molecular Biotechnology, 74: p. 257, 2001) and a fibronectin type III domain (J. Mol. Biol, 284: p. 1141, 1998). Anticolin is a capture protein altered on the basis of lipocalin. Lipocalin is a protein composed of 160 to 180 amino residues, which functions in transporting and storing substances with a low degree of solubility in water. Regarding constitution, lipocalin is a barrel structure composed of 8 β sheets. Lipocalin can recognize and bind to a substance of interest through 4 loop structures connecting the 8 β sheets. Fibronectin is generally a protein composed of amino acids not more than 100 residues, which plays a key role in the junction of cells with extracellular matrix or cell junction. Like the two proteins described above, fibronectin is a protein that has β sheet structures and recognizes a target substance through loop structures among the β sheets. A novel binding protein has been constructed by introducing random amino acid sequences into the loop structures among the β sheets of anticolin or fibronectin described above. Because those molecules have a strong molecular structure composed of β sheets in addition to a molecular recognition site, the molecules specifically recognize and bind to a substrate by the fusion to a desired functional polymeric material to thereby allow the polymeric material to be immobilized on the substrate. Moreover, the molecules are also expected to have a spacer function that keeps a certain distance from the substrate without inhibiting the desired activity of the immobilized polymeric material. However, there has been no reported case that molecule-recognizing proteins with constitutively stable scaffolds typified by the above-described antibody specifically recognize and bind to inorganic substances typified by metals or semiconductor materials.
On the other hand, in the field of detection of a variety of substances (target substances), several methods of the detection and/or quantification and the like of the target substances have been established so far, particularly for proteins such as antigens and antibodies, and sugar, lectin, and nucleic acids. For example, it is known that a labeling agent is bound to an antibody specifically recognizing and binding to a sugar or a lectin as target substances to thereby allow the detection or quantification of the target substance via the labeling agent. In general, a fine particle composed of a metal such as gold or an organic material such as latex, a fluorescent substance emitting fluorescence by excitation light in a certain wavelength region, or an enzyme having the fluorescent substance as a reaction product (e.g., HRP) is used as a labeling agent. Methods of labeling proteins such as antibodies include a method by physical absorption and a chemical bond method in which a reactive functional group is introduced into a labeling agent or a substance to be labeled and is used as a crosslinking point to form a chemical bond.
Hereinafter, prior arts will be described by taking antibodies used in detection as an example. JP 03-108115 B discloses an example of a method of labeling an antibody with gold by physically absorbing a gold fine particle onto the antibody. According to this method, a monoclonal antibody is added to a dispersion of gold colloid that has been previously adjusted, and the whole is subjected to centrifugal sedimentation to remove a supernatant liquid, and the resulting solution undergoes washing processes several times to allow the production of an antibody labeled with gold.
Next, a method of labeling antibodies by a chemical bond method will be described. Antibodies have the amino groups or SH groups of proteins. A functional group reactive with those groups is previously provided in a labeling agent to thereby allow the chemical bond between a labeling agent such as a fluorescent substance and a substance to be labeled such as an antibody. An example of the method includes a method involving introducing a labeling agent having an N-hydroxysuccinimide group, an isothiocyanate group, a nitroaryl halide group, or an acid chloride group, which reacts with an amino group. N-hydroxysuccininide that is widely used as a crosslinking agent for labeling proteins is known to efficiently react with an amino group under a pH atmosphere of 7 or more and form a highly stable amide bond (Biochemistry, Vol. 11, pp. 2291, 1972). The amino groups of the α position and of the ε position of lysine side chain on a protein can be targeted by a succinimide group in reaction. In particular, an amino group at the ε position is considered to be the general target of succinimide. For example, when a gold fine particle as a labeling agent is chemically bound to a protein such as an antibody, the gold fine particle is first modified with a compound having at least a SH group at one end and a functional group highly reactive with the above-described side chain residue of a protein at another end. Next, the protein can be crosslinked with the reactive functional group to bind them. However, because a residue having lysine or a free α-amino group nonselectively becomes a subject of interest in reaction in those methods, a protein to be labeled such as an antibody may be inhibited in its function. Although FITC is also known as a fluorescent substance having an isothiocyanate group, like succinimide, the desired property of a protein to be labeled may be reduced due to nonspecific reaction to an amino group.
Alternatively, a —SH group can be given as a crosslinking point. Methods with the use of a SH group can be broadly divided into a maleimide method and a pyridyl disulfide method. The maleimide method is a method with the use of a crosslinking agent having maleimide as a group selectively reactive with a SH group, which is known to allow selective crosslinkage under a mild condition. For example, when a subject to be bound is an antibody, a SH group that cleaved disulfide in the hinge portion of the antibody molecule is unrelated to the antigen-recognizing portion, so that the specificity of the antibody is expected not to be impaired even though the SH group is modified. This SH group is used as a crosslinking point to thereby allow the binding of a labeling agent without impairing a desired function. However, an antibody has 16 SS bonds within the whole molecule, including SS bonds for retaining the structures of a heavy chain variable region (VH) and a light chain variable region (VL) having a complementarity determining region (CDR) as the antigen-recognizing portion. Thus its function may be impaired if the reduction of the SS bond is not site-specifically performed.
JP 04-070320 B discloses a protein-labeling technique with the use of metallothionein or a fragment thereof, which is a low-molecular-weight protein capable of chelation with high affinity for a wide variety of metal ions. In this application, a technique is disclosed, in which metallothionein binds at the sulfhydryl moiety to a metal ion as a labeling agent and is bound to an antibody or the like at the other functional group given as a crosslinking point such as an amine group, a hydroxyl group, or a carboxyl group. Although sites binding to a metal ion and to a substance to be labeled are distinguished in metallothionein and can be bound with the respective substances to be bound, the binding site of the substance to be labeled is uncertain as in other crosslinking methods and problems as described above is likely to still remain.
Furthermore, means of resolving nonselective modification due to the immobilization method on a labeling agent through chemical crosslinking as described above includes a modification method by genetic engineering. It is also possible to produce a fusion protein in which a binding site is introduced into a protein by a genetic engineering approach. Known as an example is a method in which a low molecular compound biotin is chemically introduced into the end of a labeling agent such as a fluorescent substance, and the whole (strept) avidin known to bind to the above-described biotin or a biotin-binding site is introduced into the N terminal or C terminal of a desired protein by genetic engineering, which is in turn expressed as a fusion protein and bound to the labeling agent via the biotin-avidin bond. Thus, a low molecular compound is introduced into a labeling agent such as a fluorescent substance and a fusion protein in which a protein capable recognizing and binding to the low molecular compound is introduced into a desired protein is produced by genetic engineering to allow the introduction of a selective binding site.
However, the disclosed technique described above does not disclose any technique concerning a binding protein molecule-selectively recognizing and binding to the surface of a substrate material typified by an inorganic substance or a labeling agent. Therefore, when a structure in which the binding protein or a complex protein is arranged on a substrate is produced, the arrangement must be performed by a conventionally known method, for example, a method involving chemically modifying the substrate or the above-described binding protein to arrange the binding protein on the substrate through a covalent bond. Similarly, when the binding protein is bound to a fine particle of a metal or a semiconductor substance, and a labeling agent, it is also necessary to chemically modify the fine particle to be bound or the binding protein. Such chemical modification often targets amino residues or carboxyl groups contained in large numbers in the molecule of a protein, and a site involved in the reaction is nonselective. Therefore, a site exerting desired activity may be a substrate-binding or labeling site and, as a result, the desired activity of the protein may be reduced.
Moreover, such a problem is likely to arise in not only the technique of conversion into microdevices but the production of sensing elements such as biosensors. Thus, it is important to immobilize a molecule capturing a target substance such as an antibody on a substrate while molecular orientation that sufficiently exerts its capturing function is maintained.